Perfusion Imaging, Spectroscopy, and Advanced MR
Perfusion imaging quantifies tissue blood flow and helps separate infarct core (already-dead tissue) from ischemic penumbra (at-risk but salvageable). This is the central physiological question in extended-window thrombectomy and tPA decisions. CT perfusion (CTP), dynamic susceptibility contrast MR (DSC-MR), and arterial spin labeling (ASL) all generate perfusion maps with different strengths.
🔹 Bottom Line: Perfusion Imaging
- CT perfusion (CTP): workhorse for acute stroke triage; fast and widely available. Generates CBF, CBV, MTT, Tmax maps.
- DSC-MR perfusion: dynamic susceptibility contrast; brain tumor characterization (relative CBV high in high-grade glioma); also used for stroke.
- ASL (arterial spin labeling): no contrast required; quantifies CBF directly; useful in pediatric, renal-failure, or follow-up.
- Stroke decision-making (DEFUSE / DAWN): core volume + penumbra mismatch defines extended-window thrombectomy eligibility (6–24 hours).
- Core map: CBF <30% of contralateral (most labs). Penumbra: Tmax >6 sec.
- MR spectroscopy: NAA down (neuronal loss), choline up (cell turnover), lactate up (anaerobic), lipid up (necrosis). Tumor grading + radiation necrosis vs recurrence.
Perfusion Parameters
- CBV (cerebral blood volume): volume of blood per unit brain tissue (mL/100g). Falls in irreversible infarct core; preserved in penumbra (autoregulation).
- CBF (cerebral blood flow): rate of blood delivery (mL/100g/min). Falls in core; reduced in penumbra.
- MTT (mean transit time): average time blood spends in capillary bed. Prolonged in both core and penumbra.
- Tmax: time to maximum of the residue function. Sensitive to bolus delay/dispersion — prolonged in penumbra and stenosis. Most widely used penumbra marker.
- TTP (time to peak): similar to Tmax; older parameter; less robust to delay.
CT Perfusion (CTP)
Use
- Acute stroke triage — extended-window thrombectomy (DAWN, DEFUSE-3) and selected tPA decisions.
- Vasospasm monitoring post-SAH.
- Cerebrovascular reserve assessment (with acetazolamide challenge).
Acquisition
- Dynamic acquisition during iodinated contrast bolus.
- ~45–60 second scan covers a brain volume.
- Generates CBV, CBF, MTT, Tmax maps automatically (RAPID, Olea, Brainomix, vendor packages).
Core / Penumbra Definition
| Tissue |
Definition (RAPID) |
| Infarct core |
CBF <30% of contralateral |
| Penumbra (hypoperfused at-risk) |
Tmax >6 seconds |
| Mismatch volume |
Penumbra − Core |
| Mismatch ratio |
Penumbra / Core |
DAWN / DEFUSE-3 Criteria (Extended-Window Thrombectomy)
- DAWN (6–24 hours):
- NIHSS ≥10 + core <31 mL (age ≥80) OR core <51 mL (age <80).
- NIHSS ≥20 + core <51 mL (age <80).
- DEFUSE-3 (6–16 hours):
- Core <70 mL + mismatch ratio ≥1.8 + mismatch volume ≥15 mL.
Pitfalls
- Bolus timing errors: too early/too late bolus skews maps.
- Motion artifact compromises perfusion maps significantly.
- Cardiac output low: prolongs all transit parameters globally.
- Atrial fibrillation: bolus dispersion distorts maps.
- Old infarct in same territory: confounds core measurement.
- Posterior fossa: bone artifact + small volumes limit accuracy.
- Radiation dose: cumulative concern if multiple studies.
- Contrast nephropathy: risk in CKD.
DSC-MR Perfusion
Use
- Brain tumor characterization: relative CBV (rCBV) elevated in high-grade glioma; distinguishes recurrence from radiation necrosis.
- Stroke: where MRI is preferred over CT (pediatric, repeated assessment).
- Lymphoma vs glioma: lymphoma has lower rCBV than glioma despite high cellularity.
Acquisition
- Dynamic susceptibility contrast: gadolinium bolus + rapid T2*-weighted EPI imaging.
- Generates rCBV, rCBF, MTT maps.
Pitfalls
- Susceptibility artifacts near skull base, sinuses, prior surgery.
- Leakage correction required for enhancing lesions (especially gliomas, metastases).
- Gadolinium contrast required.
Arterial Spin Labeling (ASL)
Use
- Contrast-free perfusion imaging (magnetically labels arterial water as endogenous tracer).
- Pediatric (no contrast).
- Repeated assessments / cerebrovascular reserve testing.
- Stroke / cerebrovascular disease.
- Neurodegenerative disease research (regional hypoperfusion patterns).
Pitfalls
- Lower SNR than DSC-MR.
- Bolus arrival time variability in older / atherosclerotic patients can produce ATA (arterial transit artifact).
- Lower-flow regions hard to quantify reliably.
MR Spectroscopy (MRS)
Use
- Metabolic profile of a voxel.
- Tumor grade prediction: high choline / NAA ratio + lipid/lactate peaks suggest high grade.
- Radiation necrosis vs recurrence: necrosis has flat metabolic profile; recurrence has elevated choline.
- Mitochondrial disease: elevated lactate in MELAS, Leigh.
- Leukodystrophies: NAA changes; some specific patterns (NAA elevation in Canavan, NAA depletion in MLD).
- Hepatic encephalopathy: elevated glutamine-glutamate (Glx); reduced myo-inositol.
- Demyelination: NAA reduction in active plaques.
Key Metabolites
| Peak (ppm) |
Metabolite |
Meaning |
| 2.0 |
N-acetyl aspartate (NAA) |
Neuronal density (decreased in neuronal loss) |
| 3.0 |
Creatine (Cr) |
Energy metabolism (relatively stable — internal reference) |
| 3.2 |
Choline (Cho) |
Cell membrane turnover (elevated in tumors, demyelination) |
| 3.5 |
Myo-inositol (mI) |
Glial marker; elevated in Alzheimer; reduced in hepatic encephalopathy |
| 1.3 (doublet, inverts on long TE) |
Lactate |
Anaerobic metabolism; mitochondrial disease, infarct, abscess |
| 0.9–1.4 |
Lipid |
Necrosis; high-grade tumor |
| 2.0–2.5 |
Glutamate / glutamine (Glx) |
Hepatic encephalopathy; epilepsy; some neurodegenerative |
| 2.4 |
2-hydroxyglutarate (2HG) |
IDH-mutant glioma (specific!) |
Diffusion Tensor Imaging (DTI) and Tractography
- DTI measures water diffusion directionality (anisotropy) — high in white matter tracts.
- Generates fractional anisotropy (FA) and apparent diffusion coefficient maps.
- Tractography reconstructs white matter pathways from DTI data.
- Uses: presurgical planning (corticospinal tract, language tracts), MS lesion characterization, TBI (diffuse axonal injury), pediatric leukodystrophy research, schizophrenia / psychiatric research.
- Not yet a routine clinical diagnostic test; widely used for surgical planning + research.
Functional MRI (fMRI)
- BOLD (blood oxygen level dependent) signal — neural activity → local hyperemia → signal change.
- Task-based fMRI: patient performs task during scan; activated regions identified.
- Resting-state fMRI: spontaneous correlated networks (default mode, attention, language).
- Clinical use: presurgical mapping of motor, language, memory.
- Limitation: motion-sensitive; cannot replace intraoperative cortical stimulation for definitive language mapping.
🔹 Clinical Relevance: When Perfusion / Advanced MR Changes Decisions
- Acute stroke, 6–24 hours from onset: CTP for DAWN / DEFUSE-3 criteria — if mismatch present, thrombectomy benefits.
- Acute stroke, unknown onset (wake-up): DWI-FLAIR mismatch on MRI identifies treatable window for tPA (WAKE-UP trial); CTP can also identify penumbra.
- Brain tumor characterization: DSC-MR + spectroscopy distinguish high-grade vs low-grade glioma, recurrence vs radiation necrosis, lymphoma vs glioma.
- Suspected IDH-mutant glioma: 2-hydroxyglutarate peak on spectroscopy is highly specific.
- Presurgical epilepsy planning: fMRI for language + memory lateralization; DTI for corticospinal / Meyer loop visualization.
- Pediatric or contrast-contraindicated: ASL provides perfusion without contrast.
- Vasospasm post-SAH: CTP detects asymptomatic perfusion deficits before clinical decline.
- Suspected mitochondrial / hepatic / leukodystrophic: targeted MR spectroscopy.
Pitfalls and Pearls
- Core volume on CTP is an estimate: bolus dispersion in cardiac dysfunction, AFib, or low cardiac output inflates “core” — confirm with DWI when possible.
- RAPID output is widely used but vendor-specific: thresholds may not transfer directly between platforms.
- Motion ruins CTP: a single 1-second movement can void the study.
- Tmax >6 sec is the consensus penumbra threshold; stricter (Tmax >10) defines severely hypoperfused.
- DSC-MR requires leakage correction for enhancing lesions — uncorrected maps overestimate rCBV.
- ASL is degraded by atrial arrival time variability in atherosclerotic patients.
- Spectroscopy lipid peak = necrosis (suggests high-grade tumor / radiation necrosis); lactate peak = anaerobic metabolism (mitochondrial, infarct, abscess).
- 2HG peak on spectroscopy is highly specific for IDH-mutant glioma.
- fMRI cannot replace intraoperative cortical stimulation for definitive language mapping in epilepsy / tumor surgery.
- DAWN criteria require both NIHSS threshold AND core volume threshold — don’t quote them loosely.
References
- Nogueira RG, Jadhav AP, Haussen DC, et al. Thrombectomy 6 to 24 hours after stroke with a mismatch between deficit and infarct (DAWN). N Engl J Med. 2018;378(1):11-21.
- Albers GW, Marks MP, Kemp S, et al. Thrombectomy for stroke at 6 to 16 hours with selection by perfusion imaging (DEFUSE-3). N Engl J Med. 2018;378(8):708-718.
- Bivard A, Levi C, Spratt N, Parsons M. Perfusion CT in acute stroke: a comprehensive analysis of infarct and penumbra. Radiology. 2013;267(2):543-550.
- Law M, Yang S, Wang H, et al. Glioma grading: sensitivity, specificity, and predictive values of perfusion MR imaging and proton MR spectroscopic imaging compared with conventional MR imaging. AJNR Am J Neuroradiol. 2003;24(10):1989-1998.
- Choi C, Ganji SK, DeBerardinis RJ, et al. 2-hydroxyglutarate detection by magnetic resonance spectroscopy in IDH-mutated patients with gliomas. Nat Med. 2012;18(4):624-629.